linux-kernel - Re: [PATCH] usb: chipidea: udc: clear vbus_active flag in udc

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Tue, 20 Oct 2015 14:34:32 +0800
From:	Peter Chen <peter.chen@...escale.com>
To:	Jiada Wang <jiada_wang@...tor.com>
CC:	<gregkh@...uxfoundation.org>, <linux-usb@...r.kernel.org>,
	<linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] usb: chipidea: udc: clear vbus_active flag in udc_stop

On Tue, Oct 20, 2015 at 03:33:52PM +0900, Jiada Wang wrote:
> Hi
> 
> On 10/20/2015 03:28 PM, Peter Chen wrote:
> >On Tue, Oct 20, 2015 at 03:09:02PM +0900, Jiada Wang wrote:
> >>Hi
> >>
> >>On 10/20/2015 03:01 PM, Peter Chen wrote:
> >>>On Tue, Oct 20, 2015 at 11:29:18AM +0900, Jiada Wang wrote:
> >>>>Currently in udc_stop, if vbus_active flag is true, all USB activities
> >>>>will be stopped, but vbus_active flag is still left to be true,
> >>>>this causes issue, when afterwards driver tries to connect gadget
> >>>>device to host, But due to the uncleared vbus_active, some necessary
> >>>>setup steps are skipped.
> >>>>
> >>>>This patch clears vbus_active flag in udc_stop callback.
> >>>>
> >>>>Signed-off-by: Jiada Wang <jiada_wang@...tor.com>
> >>>>---
> >>>>  drivers/usb/chipidea/udc.c | 1 +
> >>>>  1 file changed, 1 insertion(+)
> >>>>
> >>>>diff --git a/drivers/usb/chipidea/udc.c b/drivers/usb/chipidea/udc.c
> >>>>index 8223fe7..b9ac228 100644
> >>>>--- a/drivers/usb/chipidea/udc.c
> >>>>+++ b/drivers/usb/chipidea/udc.c
> >>>>@@ -1762,6 +1762,7 @@ static int ci_udc_stop(struct usb_gadget *gadget)
> >>>>  	spin_lock_irqsave(&ci->lock, flags);
> >>>>
> >>>>  	if (ci->vbus_active) {
> >>>>+		ci->vbus_active = 0;
> >>>>  		hw_device_state(ci, 0);
> >>>>  		if (ci->platdata->notify_event)
> >>>>  			ci->platdata->notify_event(ci,
> >>>>--
> >>>>2.4.5
> >>>>
> >>>
> >>>This flag will be handled by vbus interrupt, would you please explain
> >>>more what problem you have met?
> >>the actual issue I had is,
> >>build kernel with USB_OTG_FSM enabled,
> >>keep usb cable plugged,
> >>load/unload USB gadget module (for example I tried g_zero or g_serial),
> >>after unload gadget module once, then the status of vbus_active is not
> >>correct,
> >>this causes host machine failed to detect gadget device when later
> >>gadget device module is loaded again.
> >>
> >>
> >
> >ci->vbus_active is only used at non-fsm mode, would you please try below
> >fix?
> >
> >usb: chipidea: otg: gadget module load and unload support
> >
> >This patch is to support load and unload gadget driver in full OTG mode.
> >
> >Acked-by: Peter Chen <peter.chen@...escale.com>
> >Signed-off-by: Li Jun <jun.li@...escale.com>
> >---
> >  drivers/usb/chipidea/udc.c | 17 +++++++++++++++++
> >  1 file changed, 17 insertions(+)
> >
> >diff --git a/drivers/usb/chipidea/udc.c b/drivers/usb/chipidea/udc.c
> >index e19d8b8..67cfa04 100644
> >--- a/drivers/usb/chipidea/udc.c
> >+++ b/drivers/usb/chipidea/udc.c
> >@@ -1755,6 +1755,22 @@ static int ci_udc_start(struct usb_gadget *gadget,
> >  	return retval;
> >  }
> >
> >+static void ci_udc_stop_for_otg_fsm(struct ci_hdrc *ci)
> >+{
> >+	if (!ci_otg_is_fsm_mode(ci))
> >+		return;
> >+
> >+	mutex_lock(&ci->fsm.lock);
> >+	if (ci->fsm.otg->state == OTG_STATE_A_PERIPHERAL) {
> >+		ci->fsm.a_bidl_adis_tmout = 1;
> >+		ci_hdrc_otg_fsm_start(ci);
> >+	} else if (ci->fsm.otg->state == OTG_STATE_B_PERIPHERAL) {
> >+		ci->fsm.protocol = PROTO_UNDEF;
> >+		ci->fsm.otg->state = OTG_STATE_UNDEFINED;
> >+	}
> >+	mutex_unlock(&ci->fsm.lock);
> >+}
> >+
> >  /**
> >   * ci_udc_stop: unregister a gadget driver
> >   */
> >@@ -1779,6 +1795,7 @@ static int ci_udc_stop(struct usb_gadget *gadget)
> >  	ci->driver = NULL;
> >  	spin_unlock_irqrestore(&ci->lock, flags);
> >
> >+	ci_udc_stop_for_otg_fsm(ci);
> >  	return 0;
> >  }
> >
> >
> 
> I will test with this patch and come back with the result,
> has this patch been submitted to community?
> 

I am afraid it is not, please add your tested-by if it fixes
your problem, I will queue it as a bug-fix.

-- 

Best Regards,
Peter Chen
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/